Tensor Decomposition for Multilayer Networks Clustering
Authors: Zitai Chen, Chuan Chen, Zibin Zheng, Yi Zhu3371-3378
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results on synthetic and real-world datasets show the effectiveness and robustness of our method against noise and irrelevant data. In this section, we conduct experiments to validate the robustness and effectiveness of CMNC against noise and multiple structures. We evaluate CMNC with 6 baseline methods on synthetic dataset, 20 Newsgroups and Digits dataset. |
| Researcher Affiliation | Collaboration | Zitai Chen,1,2 Chuan Chen,1,2 Zibin Zheng,1,2 Yi Zhu3 1School of Data and Computer Science, Sun Yat-sen University, Guangzhou, China 2National Engineering Research Center of Digital Life, Sun Yat-sen University, Guangzhou, China 3MTdata, Meitu |
| Pseudocode | Yes | Algorithm 1: Framework of the trust region method. Algorithm 2: Framework of Dogleg approach. |
| Open Source Code | No | The paper does not provide a statement or link to open-source code for the described methodology. |
| Open Datasets | Yes | We evaluate CMNC with 6 baseline methods on synthetic dataset, 20 Newsgroups and Digits dataset. We further evaluate the effectiveness of our methods using 20-Newsgroups dataset (term document frequency)... For any two graphs in the same category, a document is randomly mapped to documents in the same topic. For the graphs in the different category, the documents are randomly mapped together without considering topics. Thus, the adjacency tensor is of size 200 x 200 x 15 containing 3 groups of network and 4 document clusters for each group. ... we have the digits handwritten dataset with six hand-picked Feats: Fourier, profile, Karhunen-Love, pixel, Zernike and morphological. |
| Dataset Splits | No | The paper describes how synthetic data was generated and refers to well-known datasets, but it does not specify explicit training/validation/test splits (e.g., percentages or sample counts) needed for reproduction. |
| Hardware Specification | No | The paper does not provide any specific details regarding the hardware used for running the experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers. |
| Experiment Setup | Yes | Keeping the sparsity around 5%, we sample edges with ascending signal-noise ratio by tuning pair (α, β) from (0.1,0.03), (0.08,0.034) to (0.05,0.04). ... we set R = 2 and L = [10, 20] in which 20 is a random number for grouping the noise matrices only. |